A high accuracy survey-grade geographic information system (GIS) would need to transform distinct isolated land surveys, which could be separated by several miles, onto a common coordinate system that does not distort or scale the dimensions of those surveys. Furthermore, a high accuracy survey-grade GIS would have to position the transformed surveys relative to each other at the same distances that would be measured between them on the ground using transit and tape or electronic distance measure (EDM). Because the purpose of a high accuracy survey-grade GIS is to transform separate isolated surveys onto a common coordinate system in such a manner as to produce in essence one unified survey, in order to be survey-grade the relative positions of the transformed surveys would have to meet the relative positional accuracy standards for ALTA/ACSM land title surveys as adopted by the American Land Title Association and the National Society of Professional Surveyors, which is a member organization of the American Congress on Surveying and Mapping. Those standards state: “‘Relative Positional Accuracy’ means the value expressed in feet or meters that represents the uncertainty due to random errors in measurements in the location of any point on a survey relative to any other point on the same survey at the 95 percent confidence level. . . . [The] Allowable Relative Positional Accuracy for Measurements Controlling Land Boundaries on ALTA/ACSM Land Title Surveys [is] 0.07 feet (or 20 mm)+50 ppm.”
Global Navigation Satellite Systems (GNSS), such as the United States Department of Defense's Global Navigation System (GPS), afford land surveyors the prospect of relating all their surveys to a common spatial reference system based on geodetic latitudes, longitudes, and ellipsoid heights. In theory, the ability to relate all surveys to a common coordinate system opens the door to possible realization of a high accuracy survey-grade GIS. In practice, the hurdles and multiple problems associated with actually designing and implementing a high accuracy survey-grade GIS that can feasibly operate within a survey firm while meeting the accuracy standards leads one to conclude that such a complex system of technology married to the human management of average surveyors and field crews is at best improbable. In the past several years articles have been written and conferences have taken place that address the problem of integrating the requisite high accuracy requirements demanded of land surveys with the far less accurate spatial demands historically placed on the GIS community. The discussions have been largely talk and theorizing with no solutions proposed.
Land surveyors produce many different types of surveys or plats of survey, which are paper plots or scale drawings depicting the dimensions and orientation of a parcel of land in accordance with a written deed or legal description. A survey can include a depiction of physical man made improvements, as well as natural features, such as the topography of the terrain and vegetation. Surveyors obtain the information necessary to produce a survey by using equipment designed to measure the location of individual points on the surface of the earth.
The types of measurement equipment used may include electronic total stations and or dual frequency differential GNSS antennas and receivers that generate positional coordinates by receiving signals from U.S. Department of Defense satellites, Russian Glonass satellites, and in the future a European satellite system called Galileo. For example, if a surveyor needs to locate and dimension a roadway, he will be required to measure the relative location of a sufficient number of individual points on the edge of the road so that when those points are connected by lines or curves, the result is a correct scale rendering of the road.
As points are being measured in the field, the coordinates representing those locations may be stored in a data collector mounted to, connected to or in communication with the measuring instrument. Often, the data collected for a single point location consists of five fields within an electronic or computer point database. Those five fields, in the order most commonly used, are: 1) Point Number, often an arbitrary number automatically generated at the time of measurement and usually consecutively sequenced from the last point number used, it is used to distinguish one point from another, but may also be an assigned identifier; 2) Northing, the Y component in a three dimensional Cartesian coordinate system; 3) Easting, the X coordinate in a three dimensional Cartesian coordinate system; 4) Elevation, the Z coordinate in a three dimensional Cartesian coordinate system; 5) Point description, a code which uniquely identifies what is being located, whether it be a building corner or edge of asphalt. Other information may also be collected simultaneously or contemporaneously with these five data elements.
The electronic field measured point data may then be transferred from the data collector to an office computer of the survey company or firm and then may be imported into survey software that may be used to create a computer aided drafting (CAD) drawing that has an associated point data base with the five or more data fields as described. The CAD software may then be used to connect the dots between the points in the associated point database, based on classifications that may be included in the point description field and on input from the field crew, and may also be used to produce a plat or record of survey which may be printed out on a plotter. CAD drawings and associated point databases may be kept and managed within project folders that may include unique project numbers used to distinguish one survey from another.
The survey and description of real property in the United States has historically proceeded under the fiction that the world is flat. With very few exceptions, written legal or deed descriptions for parcels of land in the United States are based on distances that are measured on the ground in the sense that the distance between two points is measured using a tape or chain held level. Indeed this is the means by which the public lands of the United States have been surveyed and sold off to private owners beginning with the first Land Ordinance passed on May 20, 1785 by the Continental Congress: “An Ordinance for Ascertaining the Mode of Disposing of Lands in the Western Territory. Be it ordained by the United States in Congress assembled, that the territory ceded by individual states to the United States, which had been purchased of the Indian habitants, shall be disposed of in the following manner: . . . The lines shall be measured with a chain; . . . ” Legal descriptions for real property may reference adjacent or nearby land or legal features, which may be measurable on the face of the Earth as well. In the interpretation of written legal descriptions to derive a drawing or survey plat of a parcel, it is conventional to derive the location of parcel boundaries with respect to a planar or flat two dimensional Cartesian coordinate system (for the vast majority of surveys this is mandatory because almost all legal descriptions preserve a chain of title from the time they were originally conveyed by the United States). Thus, drawings or other interpretations of property descriptions are drafted from the reference of measurements upon the ground.
It might be of enormous benefit to a surveyor to be able to spatially relate, with high accuracy, all surveys he or she produced. One of the primary benefits is illustrated in FIG. 1. Depicted are four parcels of land, parcels A, B, C and D, showing the parcels' actual physical spatial relationship as measured on the ground. If surveys are produced for Parcels A, B, and C, and if the relative locations of those surveys are known with sufficient accuracy, then the amount of time and effort required to survey Parcel D could be dramatically reduced because four of the property lines of Parcel D are defined by property lines belonging to Parcels A, B, and C. The area over which a given survey or legal description for real property is likely to have influence over the location of adjacent or nearby boundaries of other parcels will generally not exceed several square miles. Of course not being able to predict which combination of surveys will have a bearing on future surveys it would be necessary to be able to spatially relate all surveys produced.
A very important characteristic of most two dimensional Cartesian systems used for legal descriptions is that they have no actual spatial relationship to each other, in many cases not even if two parcels are contiguous. In other words, given the legal descriptions of two parcels of land that are within a half mile of each other, it is not likely that their actual physical spatial relationship can be established based upon the descriptions alone. This is illustrated in FIG. 2, which depicts a possible orientation of the parcel boundaries based on deed or legal descriptions of the same parcels illustrated in FIG. 1. The orientations depicted in FIG. 2 are the orientations that must be used within the CAD drawings and associated point databases to produce plats of survey. Without a common coordinate or grid reference system which may be used to tie these disparate parcels together, the interpretation might result in the parcels “floating about in space,” as depicted in FIG. 2.
One way to establish the actual physical spatial relationship between two surveys is to measure from one parcel to the other so as to establish their relative positions. Prior to GNSS, if a surveyor wanted to determine the spatial relationship on the surface of the earth between every survey performed, the surveyor might have to physically traverse on the ground between every one of those surveys using an electronic total station, theodolite, EDM, or other suitable measuring device. Even if it were feasible to do this, it would not be possible to do so with sufficient accuracy due to the large propagation of error that would result. With the advent of GNSS and the coming on line of over 1000 Continuously Operating Reference Station (CORS) control points throughout the United States, the situation has changed with regard to coordinating and referencing different surveying jobs.
A CORS control point is a permanent fixed GPS antenna and receiver that records GPS satellite signals 24 hours a day, 7 days a week, and transmits that data as soon as it is collected to the National Geodetic Survey (NGS) where it immediately becomes available at no cost to anyone with Internet access (NGS is a branch of the National Oceanographic and Atmospheric Administration (NOAA)). The location of every CORS antenna and its electronic phase center is known and monitored with extraordinary accuracy in relation to a comprehensive continental coordinate system and datum called “NAD 83 (CORS).” The coordinates of the CORS are given in terms of geodetic latitude, longitude, and ellipsoid height defined on the WGS84 ellipsoid, a mathematical surface designed to approximate the shape of the earth. These highly accurate coordinates are down loadable from NGS websites. The network of National and Cooperative CORS constitutes the National Spatial Reference System.
An NGS Web site defines the NSRS as follows:
“The National Spatial Reference System (NSRS), defined and managed by the National Geodetic Survey (NGS), is a consistent national coordinate system that specifies latitude, longitude, height, scale, gravity, and orientation throughout the Nation, as well as how these values change with time.”
“NSRS consists of the following components:                A consistent, accurate, and up-to-date National Shoreline;        the National CORS, a set of Global Positioning System Continuously Operating Reference Stations meeting NOAA geodetic standards for installation, operation, and data distribution;        a network of permanently marked points including the Federal Base Network (FBN), the Cooperative Base Network (CBN), and the User Densification Network(UDN); and        a set of accurate models describing dynamic geophysical processes affecting spatial measurements.”        
“NSRS provides a highly accurate, precise, and consistent geographic reference framework throughout the United States. It is the foundation for the National Spatial Data Infrastructure (NSDI), a critical component of the ‘information superhighway.’ NSRS is a significant national resource—one whose value far exceeds its original intended purpose.”
Surveyors may use an extremely accurate type of positioning utilizing GPS, known as dual frequency relative positioning, which requires that two or more GPS receivers operate simultaneously receiving and recording satellite data from common satellites. With the two or more GPS receivers operating simultaneously and receiving signals from common satellites, the satellite data recorded by the receivers can be downloaded to a computer and post-processed using software designed for that purpose (GPS that utilizes post-processed vectors is called static GPS). The result is a highly accurate vector within WGS84 defining the relative position of the two GPS antennas. Very importantly, if the absolute position of one of the antennas is known and held fixed within the NSRS, then the vector derived from post-processing is no long relative and determines the absolute position of the second antenna or point.
When surveyors use dual frequency relative positioning GPS, one of the two GPS antennas is usually called a base station and remains positioned over a control point in the ground for many hours at a time, sometimes over successive days. The other antenna and receiver is called the rover and is moved from point to point with short occupation times in order to establish real time kinematic (RTK) GPS vectors or post processed static GPS vectors relative to the base station. If, in addition to deriving RTK and or static vectors between the base station and rover, vectors are also derived between the base station and one or more CORS through static post-processing, then highly accurate absolute positions for both the base station location and the points located by the rover relative to the base station can be computed within the NSRS. Because many large survey firms now employ GPS routinely in connection with most of their surveys, it may be possible for them to practically establish the absolute (within the NSRS or some other encompassing coordinate system) and therefore relative positions of those surveys to a very high degree of accuracy. Any measurement errors in the vectors from three or more CORS to the base station can be adjusted, for example by the method of least squares, holding the published CORS coordinates fixed. Such an adjustment computation may result in positions for the base station, and the associated points within a particular survey job, that exceed in accuracy the positions that could be achieved through the use of conventional traverses run by using electronic total stations and tying the surveys to conventional ground control stations. These higher levels of accuracy can be achieved virtually every time with generally two hours of observation at the base station by post processing base station GPS data with CORS control point data that has been downloaded from NGS websites. The CORS data may have been collected hundreds of miles from the base station and the site of the survey.
A few states in the United States have what are called virtual reference systems (VRS). Europe is blanketed by such systems. A VRS is a network of CORS that immediately relay their data to a central computer that then models the atmospheric corrections over the area encompassed by the network. These atmospheric corrections are then conveyed via cell phone to GPS rovers operating in the field. The result is real time or RTK positions at the rover without the need for a base station set up near the site of the survey. In the United States VRS systems are all operating on the NSRS and NAD 83 (CORS). Therefore a surveyor who is operating in a VRS is automatically establishing a link between local survey points and an encompassing coordinate system, in this case the NSRS.
In order for GPS located points to be usable for spatially relating unconnected surveys in a high accuracy survey-grade GIS their WGS84 latitude and longitude coordinates must be transformed into grid coordinates by defining a map projection. The term “grid” refers to a Cartesian coordinate system that is the result of a map projection. A map projection projects points on a curved surface onto a conical or cylindrical three dimensional surface which can be cut and laid flat, thereby transforming coordinates for points located in three dimensions on a curved and irregular surface into points represented in a flat two dimensional frame. A map projection typically includes an ellipsoid designed to approximate some aspect of the earth's surface (such as, but not limited to, mean sea level) and a conical or cylindrical surface passing through or around the ellipsoid onto which points on the surface of the earth are projected. From a simple geometric standpoint that can be visualized, a projection can be accomplished by projecting lines from the center of the ellipse through points on the surface of the earth (see FIGS. 3, 5). Where the lines intersect the conic or cylinder defines the location of the points in the grid system when the conic or cylinder is cut and laid flat. In most practical applications a map projection is a mathematical operation defined by functions that relate geodetic latitudes and longitudes in a spherical system to X and Y coordinates in a two dimensional Cartesian grid system.
The tradeoff for representing on a flat surface the relative size, shape, and location of figures that exist on a curved surface is that the correct shapes and distances as they exist on the curved surface become distorted on the flat surface. This is evident to anyone who has seen a flat map of the world and noticed that Greenland appears to be larger than the continental United States. The larger the area of the earth depicted using a map projection, the greater the distortion. The converse is also true, as the area of the earth encompassed by a map projection becomes smaller so to can the distortion. Because the areas over which it may be desirable to spatially relate surveys is on the order of several square miles, it becomes possible to design map projections that reduce the difference between grid distances and ground distances to an order well within the measurement tolerances associated with the best practices of land surveying.
Because the coordinates that are produced using GPS are in terms of latitudes and longitudes, which are defined in a three dimensional spherical frame, these coordinate systems cannot be used as a basis for spatially relating legal descriptions which are defined within two dimensional Cartesian coordinate systems, as are required in the development and processing of local land surveys. The local land surveys are typically referenced to a locally optimized coordinate system and may be arranged so that a computed grid distance and a measured ground distance are within an acceptable level of tolerance for any location where the local coordinate system may be used.
It is desirable that improvements to the processing of coordinates for disparate surveying jobs in a particular geographic area be made so that surveys of different origins and dates can be compared and harmonized with each other.